molecules

Review
Cortisol Monitoring Devices toward Implementation for
Clinically Relevant Biosensing In Vivo
Pavel A. Kusov 1, *, Yuri V. Kotelevtsev 2 and Vladimir P. Drachev 1

1 Center for Engineering Physics, Skolkovo Institute of Science and Technology, 121205 Moscow, Russia
2 Vladimir Zelman Center for Neurobiology and Brain Rehabilitation,
Skolkovo Institute of Science and Technology, 121205 Moscow, Russia
* Correspondence: p.kusov@skoltech.ru

Abstract: Cortisol is a steroid hormone that regulates energy metabolism, stress reactions, and im-
mune response. Cortisol is produced in the kidneys’ adrenal cortex. Its levels in the circulatory system
are regulated by the neuroendocrine system with a negative feedback loop of the hypothalamic–
pituitary–adrenal axis (HPA-axis) following circadian rhythm. Conditions associated with HPA-axis
disruption cause deteriorative effects on human life quality in numerous ways. Psychiatric, car-
diovascular, and metabolic disorders as well as a variety of inflammatory processes accompanying
age-related, orphan, and many other conditions are associated with altered cortisol secretion rates and
inadequate responses. Laboratory measurements of cortisol are well-developed and based mainly on
the enzyme linked immunosorbent assay (ELISA). There is a great demand for a continuous real-time
cortisol sensor that is yet to be developed. Recent advances in approaches that will eventually
culminate in such sensors have been summarized in several reviews. This review compares different
platforms for direct cortisol measurements in biological fluids. The ways to achieve continuous
cortisol measurements are discussed. A cortisol monitoring device will be essential for personified
pharmacological correction of the HPA-axis toward normal cortisol levels through a 24-h cycle.

Keywords: cortisol continuous monitoring; in vivo biosensor; neuroendocrinology; nanomaterials

Citation: Kusov, P.A.; Kotelevtsev,

Y.V.; Drachev, V.P. Cortisol
Monitoring Devices toward 1. Introduction
Implementation for Clinically
Cortisol monitoring devices, corticometers, designed for the long-term measure-
Relevant Biosensing In Vivo.
ments of cortisol concentration fluctuations in vivo, in interstitial fluid, and blood plasma
Molecules 2023, 28, 2353. https://
are in high demand for drug development and the pharmacological correction of the
doi.org/10.3390/molecules28052353
HPA-axis [1,2]. However, the creation of a biosensor matching all of the requirements of
Academic Editors: Wenbo Lu and biomedical science to such instruments is challenging enough. Cortisol complex pharma-
Xiaowei Cao cokinetics, its chemical and biological properties, require an analytical strategy aimed at
avoiding all the error sources that affect cortisol concentration monitoring in biofluids in a
Received: 11 February 2023
Revised: 23 February 2023
time-resolved manner [3].
Accepted: 24 February 2023
Laboratory measurement of cortisol concentrations in the physiological range of
Published: 3 March 2023 tens to hundreds of nanomol/L—in human blood, saliva, and urine—can be achieved by
immunofluorescent and other commercially available techniques including ELISA and mass
spectroscopy [4]. However, this requires bench-based equipment and trained technicians
to perform, and the samples are taken and analyzed in a discrete manner, disturbing the
Copyright: © 2023 by the authors. subjected patient and introducing more error.
Licensee MDPI, Basel, Switzerland. A number of techniques have been created and applied to measure cortisol with
This article is an open access article various sampling techniques in sweat, saliva, tears, urine, however, none of them are
distributed under the terms and
efficient enough to be clinically significant or practical enough to be presented on the
conditions of the Creative Commons
market and adapted to monitoring and diagnostic devices [5–7]. Additionally, the tissue
Attribution (CC BY) license (https://
levels of cortisol are affected by cortisol to cortisone interconversion by 11β hydroxysteroid
creativecommons.org/licenses/by/
dehydrogenase (11βHSD) enzymes (Figure 1). 11βHSD type 2 is responsible for the
4.0/).

Molecules 2023, 28, 2353. https://doi.org/10.3390/molecules28052353 https://www.mdpi.com/journal/molecules

Molecules 2023, 28, 2353 2 of 21

conversion of cortisol to inactive cortisone in mineralocorticoid sensitive tissues, while

type 1 regenerates cortisol from cortisone in the liver, brain, and some other tissues [8].
The activity of cortisol-converting enzymes in salivary and sweat glands are known to be
sources of error, which can be identified by comparing the immunofluorescent assay results
with the mass-spectroscopic ones [9].

Figure 1. Cortisol–cortisone 11βHSD isoenzyme 1–2 “shuttle”.

The activity of the 11βHSD enzymatic shuttle system is currently a target for the
development of isozyme-specific inhibitors, preventing inadequate overactivation of the
stress response mechanisms under conditions associated with high local cortisol levels such
as intracranial hypertension [10,11]. Cortisol molecules are lipophilic and overcome the
blood–brain barrier. This steroid exists in blood serum in a dynamic equilibrium of its bio-
logically active free form and biologically inactive portion, selectively reversibly bound to
transcortin (corticosteroid binding globulin—CBG) and non-specifically to serum albumins
(due to albumin vast quantity compared to CBG concentration in blood). Currently, the
quantitative analysis of cortisol in the blood and serum samples in vitro are affected by the
facts of the CBG-bound and free cortisol dynamic equilibrium in the sample, temperature,
and CBG expression rates in tissues. Some of these issues are currently being addressed
by several recalculation techniques—Coolens, Södergård, Dorin, and Nguen equations.
However, Molenaar et al. demonstrated the low clinical significance of such measurements
and the recalculation results of critically ill patients as well as non-critically ill patients [12].
Authors have commented on such findings by pointing out the fact of the affinity change
of CBG due to the NE cleavage of the reactive loop and genetic variations of the CBG gene,
decreasing the CBG protein affinity to cortisol.
CBG has one binding site for cortisol with a high (nanomolar) affinity to cortisol (Kd of
33 × 10−9 M), being found on the cellular membranes in renal tissue, liver, and others. Some
proportion of the total cortisol (up to 5%) is bound to blood albumins non-specifically [13].
The CBG saturation concentration of cortisol is reported to be at the peak level of cortisol at
approximately 400 nM/L [14]. The free cortisol portion (up to 5% of total) is biologically
active and its concentration fluctuates with certain frequency due to the adrenals’ pulsatory
release of cortisol and the CBG–cortisol dynamic depo buffers the influx of hormones into
the circulation and target tissues [15]. CBG is possibly involved in the membrane transport
of steroids as it was found in the cellular cytoplasm. The spatiotemporal biodistribution
of free cortisol is regulated by the local activity of specific proteases—neutrophil elastases
during acute trauma, inflammation, or sepsis. CBG expression rates are also important
features and regulators of the cortisol complex biodistribution [16,17].
Thus, cortisol is locally enzymatically inactivated and reactivated in sweat and saliva,
making the data obtained by non-invasive sampling less accurate and deceiving, while
blood puncture itself is a stressful event triggering cortisol release and waking during the
sleep hours. Exocrine gland (sweat, saliva, tears—cortisol sample sources) channels are
natural dialysis systems excreting only the free cortisol component of the total cortisol as
blood carriers are cut off. Due to this, the proportion of cortisol concentration in saliva to
that of a serum is different. As a result, biosensors aimed at excreted sample sources will
Molecules 2023, 28, 2353 3 of 21

require calibration with parallel measurements of cortisol in the blood or the interstitial
fluid (ISF) samples.
High variation can be found in cortisol concentrations measured in the same samples
using immunoassays and mass-spectrometry. The hormone concentration is best measured
in the specific vascular bed or at the site of its action. This stimulates the development of an
implantable cortisol biosensor [18,19]. A biosensor structurally consists of a biorecognition
element, which upon binding the analyte molecule, initiates a physico-chemical signal
registered by an acquisition element and is transmitted to the receiver. The signal calibration
adjustment results in a graphical or digital representation of the analyte concentration in
the sensing volume. High interest and demand for the implantable spatiotemporal probe
of free (active) cortisol concentrations yielded a plethora of analytical systems potentially
applicable for the stated goals.
Nevertheless, there is no medical-grade analytical device present to be used as a
continuous cortisol monitoring device (CCM) to support normal hormonal allostasis and
prevent acute cortisol-related conditions [20]. The existing continuous glucose monitoring
(CGM) devices and their clinical significance are reviewed in the Section 2.1. It serves as an
example for the development of the potential CCM device prototype.

2. Continuous Monitoring Device

2.1. Example of the Clinically Relevant, FDA-Approved Long-Term Implanted Concenration
Monitoring Medical Device: Continuous Glucose Monitoring (CGM) Eversense
The Seneonics Eversense implanted glucometer consists of an implanted, for up to
180 days, sensing probe connected wirelessly with a matchbox size electronic transmitter,
patched to the skin by tape just above the site of the injected (under the skin) 2 cm length
polymer capsule (see Figure 2). This is a successful example of such an analytical system
and has been proven to be highly efficient in accurate insulin replacement therapy [21].

Figure 2. Scheme of the Senseonics Eversense implanted module probe containing one or two
(depending on the model—for 90 or 180 days) silicon collars slowly releasing dexamethasone to
diminish the inflammatory response in the implantation site. The indicator polymer is in proximity of
the optical system that flashes the UV-LED to measure the fluorescence intensity of the indicator gel.
The controller performs the measurements and connects to the externally worn transmitter wirelessly
to collect and store data points. From [22] with permission from Wiley.

The sensor probe is a PMMA capsule containing an analytical system equipped with a
medical grade silicone tube secreting anti-inflammatory medicine to protect the capsule
from the immune reaction. A small-molecule permeable cap as a semi-membrane allows
glucose to passively diffuse to ensure its contact with the sensitive polymer. The indicator
component of the gel—polycyclic aromatic molecules—specifically and reversibly bind
glucose molecules with a structural reconfiguration allowing the fluorescent signal to be
generated by the glucose–indicator complex, as shown in Figure 3. Thus, the fluorescence
Molecules 2023, 28, 2353 4 of 21

intensity is proportional to the glucose concentration. Indicator molecules are, in fact,

side-branches of the polymer immobilized on the surface of the device sensing area [23].

Figure 3. Equilibrium binding of glucose to the indicating polymer. R1 and R2 represent the polymer
chain terminals’ “upstream” and “downstream” depicted point. From [22] with permission from Wiley.

The capsule is surgically implanted in a patient’s shoulder, belly, or back with a

special injection needle by trained medical personnel. A transmitter module is wearable
by patching it to the skin over the implanted capsule to perform daily data collection
that is represented by a smart watch and smartphone application via Bluetooth. The
transmitter reads data from the implanted probe, automatically calibrates and analyzes it,
and then stores the data, sending it to the external devices to be read eventually by the user.
Further analysis of the long-term monitoring data is performed by a physician to correct
the medicine administration regimen and dosage. It was evidentially demonstrated that the
usage of a continuous implantable glucose monitor is safe and beneficial to the efficiency of
the insulin replacement therapy as well as to the overall functional improvement in the
patients’ quality of life [24].
Eversense’s elegant decision to utilize the biorecognition element simultaneously as a
signal generation molecule is realized through the equilibrium binding of glucose to the
indicator group. Reversibility, selectivity, and fast enough kinetics of the reaction are the
keys to the creation of a successful operating time-resolved analytical system. With the first
generation CGM Eversense (90 days of safe implantation period), users were required to
recalibrate the device each day twice with a finger puncture for blood sampling. The second
generation does not longer require daily recalibrations and has two units of inflammation
suppressing dexamethasone collars on the implanted probe capsule. It has an extended
safe implantation period of up to 180 days, which demonstrates the high practical value
and usability of implantable CGM devices to the end user.

2.2. Demands of the Continuous Cortisol Monitoring Device in Clinics, Point of Care, and
Drug Development
Improvement in the quality of life of people with neuroendocrine conditions through
daily monitoring of a biomarker molecule is achievable with a biosensor, as has been
used successfully for insulin replacement therapy. An in vivo biosensing analytical sys-
tem (performing as required by the biological aspects of the task parameters) could also
be developed to assess the reactive hypothalamo–pituitary–adrenal (HPA-axis) system
of corticoid function regulation with an implantable biosensor for continuous cortisol
monitoring (CCM).
Molecules 2023, 28, 2353 5 of 21

There is still no example of a functional, practical, clinically significant, FDA-approved,

or under ongoing preclinical study of such an implanted corticometer. However, there
is very active research in this field with a number of publications reviewing this topic,
annually multiplying the number of articles on novel cortisol biosensor designs, and
discussing the future directions on continuous biosensing development [25].
Conditions associated with HPA-axis disruption are disbalancing the normal circa-
dian cortisol secretion pattern and responses with a number of diminishing life quality
physiological effects that affect the central nervous system, synaptic plasticity, cognition,
and memory [26–28].
Chronic inflammation associated with tissue degenerative processes of disease and
aging such as arthritis, multiple sclerosis (MS), atherosclerosis, asthma and often genetic
disorders (for instance, Duchenne muscular dystrophy (DMD)) are treated with systemic
long-term synthetic corticosteroid medicines like prednisone, deflazacort, and dexametha-
sone [29]. Such treatment could dramatically influence the HPA-axis, causing Cushing-like
adverse effects—fatigue, bone fragility, metabolic, circulatory, immune, and nervous sys-
tems disruptions [30].
Such patients may benefit from clinically relevant devices aimed to diminish the
adverse effects of pharmacotherapy and prevent critical conditions at the point of care.
A continuous cortisol monitoring device will help the therapist to analyze the long-term
collected data of the users’ HPA-axis response on the change in drug type, combinations,
regimens, and dosages [31–33]. The goal of such monitoring and therapy is to set the
circadian and ultradian rhythms by means of chronotherapy to integrate endogenous
cortisol rhythms.
At the moment, pharmacological corrections are prescribed based on less accurate
biofluid sampling, dialysis system connection, and technically highly precise mass spec-
troscopy and tandem mass spectroscopy-chromatography (LS-MS). Such methods tether
the patient to heavy, sophisticated laboratory equipment for 24 h or more for monitoring.
It may introduce additional errors through the discrete sampling, blood puncture, which
disturbs the patient, releasing additional cortisol, and disrupting patients’ sleep at night.
Currently, cortisol daily profiling is performed in diagnostic laboratories as a point-by-
point collection of the serum samples for conventional immunoassays or mass-spectrometry
methods to measure the free cortisol concentration in vitro. Blood puncture for each-
hour determination of cortisol, with further LS-MS measurements or ELISA, requires
sampling invasions, influencing the studied patient neuroendocrine responses. Other
sample sources such as tears and sweat are less disturbing as they require no sampling
procedures. However, these techniques are not practical enough since the patches have no
reliable adhesion to the skin. Cortisol exists in urine in free (unconjugated) and conjugated
(e.g., glucuronide-conjugated and sulfate-conjugated) forms. Approximately 1–2% of
protein-unbound circulating cortisol is excreted in the urine. Therefore, the level of cortisol
in the urine to some point reflects the level of the protein-unbound (biologically active or
free) form of cortisol in the plasma. Cortisol-converting enzyme activity in the sweat, tears,
and saliva make the results of such measurements less reliable. Thus, the most accurate
sample sources are blood, serum, or interstitial fluid.

3. Perspective Cortisol Biosensors Developed towards Continuous Monitoring:

Recognition Elements, Physico-Chemical Phenomena for Signal Generation and
Analytical Strategies
3.1. Electrochemical Cortisol Sensing
Electrochemical methods include various signal generation techniques based on the
measurement of standard parameters of the circuit—potentiometry, amperometry, im-
pedimetry, and voltammetry.
Non-enzymatic electrochemical cortisol assays can be an interesting direction for CCM
biosensor development with the same element functioning as the recognition and the
signal generation component. The assay was built on the recent findings of the steroid C3
Molecules 2023, 28, 2353 6 of 21

keto-group being prone to reduction in a preferable manner if its carbonyl is in conjugation

with a double bond [34] (Figure 4).

Figure 4. Cortisol reduction is preferable at the C3 carbonyl position of cortisol. Reproduced from [35]
under CC BY-NC 3.0.

In [34], by formulating a highly efficient reducing substrate of a ZnO layer on graphene

oxide, the electrode was created to be sensitive to cortisol in the nanomolar range of
concentrations. The described sensor performed in vitro and operated with real saliva
with only a 5% error in comparison to the artificial saliva samples calibration. While, this
technique was not further applied toward measurements in the blood or ISF, it gives an
interesting example of an electrochemical CCM biosensor since no enzyme or specific
biorecognition element is used. However, known organic substances selectively reducing
cortisol at C3 are not efficient enough in terms of the reaction yield, so further research
and optimization could be beneficial for the development of more selective electrochemical
sensors for cortisol [35].
Most cortisol biosensors are built on redox or electrochemical changes in the microen-
vironment of the probe’s surface modified with biorecognition elements. Some strategies
have been applied to amplify and transmit a binding event such as resistance change
through additional components including structure-switching aptamers, enzymes, and
competitive immunoassays with a secondary reporting tracer [36].
Another interesting strategy exists for the electrochemical biosensing of various ana-
lytes including cortisol with a biocatalytic reaction of electrode-linked enzyme (oxidases or
dehydrogenases are used: as the enzyme catalyzes its reaction, the coenzymes required
(NAD+, NADP+, NADH, NADPH, ATP FAD, FADH) are enzymatically modified to be
detected and measured by amperometry [37].
The most important obstacle in the way of electrochemical cortisol biosensors in the
development of in vivo applications is the sensitivity of the developed systems to the
pH and redox potential of the sample matrix. This issue could be solved with sample
preparation, prior measurements by filtration, or the introduction of complex membranes
selective enough to protect the sensor from the environment, yet still permeable for the free
diffusion of amphiphilic cortisol molecules with physiologically relevant rates.
Recent publications have demonstrated an active process of the search for an effi-
cient protective membrane for lipophilic analytes such as steroid hormones [38]. Here,
lipophilization of the track membrane (with low selectivity) was performed for both sensor
protection from fouling by the biological matrix of the sample and from the pH and redox
factors of the medium.

3.2. Immunosensors for CCM

A surface-specific time-resolved plasmon-enhanced fluoroimmunoassay (PEF) was
developed for cortisol concentration changes over time measurements that aimed to demon-
strate the applicability of the novel analytical signal generation technique for the continuous
cortisol monitoring task [39]. The system has a limit of detection (LOD) of 0.02 µg/mL and
responsivity of about 30 min (the reaction time decreases down to 20 min when performed
at 36 ◦ C—normal body temperature). The system can be further transferred to the optical
fiber terminal side surface and covered with a polymer capsule with a porous selective
membrane to be implanted for in vivo tests in CCM device development. The system’s
Molecules 2023, 28, 2353 7 of 21

LOD and responsivity is sufficient to cover the normal cortisol levels in blood, serum, and
saliva [3]. The application of the PEF was proposed to be formulated on the surface of the
optically active modified surface regions of terminally decladded optical fibers (sensing
region) connected through a signal transmission region (the rest of the fiber with intact clad)
of the optical fiber connected to the spectrometer and computer through optical connectors.
Such back-collecting fiber probe spectroscopy is under development for the CCM task
with the already functioning in vitro PEF time-resolved assay. This technique could be a
practical, relatively simple, and low-cost to produce sensitive CCM device that is able to
read the cortisol fluctuations in the tissue of interest. Note that the proposed probe has a
miniature size with a diameter up to 500 µm and length less than 1 cm.
The analytical system consists of the monoclonal antibodies to cortisol, raised against
the 3-HS-BSA immunogen with nM affinity to free cortisol. The antibodies are bound to
the thin plasmon-supporting film consisting of Au nanostructures on the SiO2 chip surface
deposited by physical vapor deposition processes (Figure 5).

Figure 5. (a) SEM image of the plasmon-active surface—Au nanostructured discontinious film on the
SiO2 chip surface (the task of the time-resolved cortisol monitoring relies on the plasmon enhanced
fluorescence phenomenon, which is dependent on the distance between the fluorescence reporter and the
surface in the range of tens of nanometers). (b) Here, the Au-modified surface demonstrated the ability
to generate an analytical signal through 100-fold enhancement of the red (λex = 633 nm, λem = 655 nm)
fluorescent reporter intensity in the surface immunocomplex compared to the same immunocomplex
fluorescence intensity on the polystyrene well plate (from [39] under CC BY-NC-ND 4.0).

The introduction of a competitive displacement assay scheme with surface-bound

antibodies to cortisol and bulk mobile reporter element—bovine serum albumin (BSA)—
conjugated covalently with cortisol hapten (cortisol-3-HS or cortisol-3-CMO) and fluo-
rophore molecules (FBC) is proposed. Such a scheme allows for recording, with a lag
of about 20 min, of the reversible equilibrium binding of the FBC competing with free
cortisol for the surface-bound antibody binding sites as the FBC fluorescence intensity is
maximal when no free cortisol is present. The FBCs are bound to the antibodies and in
close proximity to the Au nanostructured surface. The fluorescence intensity of the system
is inversely proportional to the free cortisol concentration in the system. The device can
measure the free cortisol concentration variations in the system as FBC is displaced out of
the antibody binding sites by free cortisol molecules and passively diffuse out of a ~15 nm
(antibody linear size) distance from the Au particles, leaving a surface enhancing area
(Figure 6).
Molecules 2023, 28, 2353 8 of 21

Figure 6. Competitive equilibrium binding of cortisol to the surface-bound monoclonal antibodies

to cortisol. Fluorescence intensity of the cortisol–BSA–fluorophore reporter is enhanced in close
proximity to the Au nanostructured film surface. When free cortisol is present, the fluorescence
intensity gradually drops, allowing the concentration fluctuations to be monitored over time in vitro
(from [39] under CC BY-NC-ND 4.0).

The other example of a cortisol immunosensor, and simultaneously cortisol aptasensor,

is based on the interaction of surface-bound quantum dots with immobilized antibodies
or aptamers that bind cortisol. Upon binding, fluorescence quenching is detected—with
an intensity proportional to the bound cortisol concentration—that allows one to measure
the cortisol concentration in saliva [40]. Such an immunoassay, being sensitive enough
to measure the saliva cortisol concentration changes (whose range in nM/L is less than
that in blood or serum—0–14 nM/L), could be further applied to the task of real-time
cortisol concentration measurements if further developed. However, the demonstrated
assay aims for a less clinically relevant but still practical enough salivary cortisol “one at a
time” point measurement.
Another cortisol immunosensor is based on the novel surface substrate, a partially
decladded polymer optical fiber (POF) covered with a continuous Au thin film for SPR
generation, which demonstrates the applicability of a novel surface and signal transmitting
element (optical fiber) for cortisol monitoring. However, the method requires a fiber to be
connected to the spectrometer and a light source from both ends, but the sensitive area
is suitable for use in vitro for low-volume microfluidic measurements as it is not quite
practical to use such fibers to develop implantable devices for in vivo monitoring.
POF mechanical properties are beneficial for in vivo application since its core material—
PMMA—makes the hair-thin waveguides flexible and biologically inert as well as prone to
covalent modification with various nanochemistries to functionalize its surface. However,
even a flexible material such as this is not durable enough to be strained in a loop toward
the in vivo measurements by SPR; the fiber loses its waveguide properties and mechani-
cally degrades when sharp angles are introduced to the fiber by bending in such a loop.
The spectra of the optical assays were adapted to be used on the tip and side surface of
the terminals of the fibers, attracting attention for implantable device development with
polymer optic fiber technology [41,42].
The first original study describing the continuous monitoring of cortisol in samples
obtained by the microdialysis of blood plasma was recently published [43]. The registration
of plasma cortisol interaction with the antibody receptor is achieved by the biosensing of
particle mobility (BPM) affected by single-molecule binding and unbinding events [44].
Polymer particles (1 µm in diameter) were functionalized with monoclonal antibodies
to cortisol (Abcam G53) This antibody was selected from several commercially available
monoclonal antibodies to balance high affinity with the reversibility of binding to the
antigen. The particles decorated with G53 antibodies were tethered by long dsDNA to the
Molecules 2023, 28, 2353 9 of 21

substrate covered with cortisol-analogues attached to the surface through a short dsDNA
linker (Figure 7).

Figure 7. Competitive equilibrium binding of the free cortisol and cortisol hapten bound with a
DNA linker to the polymer on the surface. Here, the cortisol concentration influences the mobility of
the microparticles bound to the surface—when cortisol is abscent, microparticles are bound tight to
the surface, and free cortisol displaces the surface-bound by the short DNA linker cortisol hapten,
which could be detected by the analysis of the mobility of the microparticles by optical methods [43].
Reprinted with permission of ACS (https://pubs.acs.org/doi/10.1021/acssensors.2c01358).

The density of the G53 antibody on the particles was optimized to allow for interaction
through a single specific molecular contact between the antibody molecule attached to
the particle and a cortisol analogue attached to the substrate. Reversible antigen antibody
interaction resulted in binding and unbinding events between the particles, still tethered to
the substrate by the long dsDNA molecule. The frequency of antibody mediated binding
events was reduced when cortisol was present in the solution. Bound and unbound states
of the particles were detected by recording the Brownian movements of 2000 individual
particles in the field view of a contrast microscope. It was shown that the sensor responds
to cortisol in the high nanomolar to low micromolar range and can monitor physiological
plasma cortisol concentrations over 6 h. Displacement curves obtained by increasing and
decreasing free cortisol concentrations almost coincided with an IC50 close to 1 µM. This
allowed for measurements of the 10 nM increments of cortisol concentration within 10 min
intervals, which is suitable for the continuous measurement of physiological fluctuations of
the hormone concentration in the plasma. This experiment proved the feasibility of using
antibodies as a primary receptor for continuous measurement of the physiological cortisol
concentrations. The main limitation of this method is the necessity of the dialysis of plasma
as plasma proteins will apparently increase the nonspecific adhesion of the particles to
the substrate. Stability of the antibody, particularly of the dsDNA linkers in the biological
environment, may also cause a problem. Additionally, it seems to be difficult to adapt the
BPM method based on contrast microscopic registration to the portable registration device.
Still, it is hard to overestimate the significance of this paper as it proves the possibility of
the development of a continuous cortisol sensor based on an antibody receptor.
Undoubtedly, great success has been achieved in measuring the concentration of cor-
tisol on the surface of the skin in perspiration or in interstitial fluid in the subcutaneous
layer using portable sensors [9,45,46]. Antibodies and aptamers were used as the primary
receptors. The sensors consist of microfluidic chambers and microchips that allow the
surface electrochemical phenomena or lateral transfer to be measured. Thus, in one of the
earliest works, measurements were carried out in samples of interstitial fluid collected
for 6 h through channels formed by a laser scalpel. The photolithographic technique was
used to fabricate an oxidized silicon wafer in a clean room environment. Gold microelec-
trode arrays were functionalized with non-specified anti-cortisol (Mab) using a dithiobis
(succinimidyl propionate) (DSP) self-assembled monolayer resulting in an ultrasensitive,
Molecules 2023, 28, 2353 10 of 21

disposable, electrochemical cortisol immunosensor using the electrochemical impedance

(EIS) technique. This sensor was reusable, but not continuous and not reversible. How-
ever, it demonstrated the possibility of measuring the physiological concentrations of
cortisol with a nM resolution using the EIS technique in combination with an antibody
receptor immobilized on the gold microelectrodes. The sensitivity was further increased
with functionalized antibodies specific to cortisol of the conductive carbon yarn (CCY)
with ellipsoidal Fe2 O3 [46]. Cortisol was measured in the collected sweat samples with
cyclic voltammetry (CV), or a so-called electronic tongue, which is a powerful and popular
electrochemical technique commonly employed to investigate the reduction and oxidation
processes of molecular species [47]. A prototype of a wearable cortisol sensor based on a
cortisol specific aptamer (a specifically designed oligonucleotide with high affinity binding
to the ligand) and a platinum/graphene extended gate field effect transistor (EG-FET) as a
detection method was recently described [48]. The measurements were conducted in the
solution in vitro. However, the device shows promising parameters for the real-time moni-
toring of the cortisol in human sweat. Specifically, in the solution, hysteresis-free EG–FET
curves were recorded from a 1 nM to 10 µM cortisol concentration with a detection limit of
0.2 nM with a negligible drift over time and high selectivity. However, low volume and a
slow flow rate of the sweat might be an obstacle to the reversible continuous monitoring
of cortisol concentration in vivo. Analytical performance, sample sources and analytical
signal generation techniques of the recently developed immunosensors are summarized in
Table 1.

Table 1. Immunosensors.

Sample Source Linear Range, LOD Signal Generation Reference Applicability In Vivo
104
0.001 ß to nM Field effect transistorm
Saliva [49] No
LOD (–) (Liquid Gate, Graphene)
1 to 400 ng/mL Magnetically assisted SERS
Urine and Serum 2.75 µM–1 mM immunoassay [50] No
LOD 3 ng/mL (MA-SERSI)
Serum, blood 50–200 ng/mL
(137 µM–0.5 mM)
Saliva 1–40 ng/mL Non-faradaic electrochemical
Universal (2.75 µM–0.1 mM) impedance spectroscopy [51] No
Sweat 10–150 ng/mL (EIS)
(27 µM–0.4 µM)
LOD 1 ng/mL
156~10,000 pg/mL
Chemiresistor graphene
Saliva (0.4–2.7 µM) [52] No
oxide sensor
LOD 10 pg/mL
Dialyzed
100 nM–10 µM Yes; Pre-microdialysis
reconstituted Particle mobility assay [43]
LOD (–) required
Human plasma
1 pM to 100 nM Electrochemical impedance Yes; ISF extractor
Extracted ISF [53]
LOD (–) immunoassay required
Chronoamperometry and
1 pg/mL to 150 ng/mL
non-faradaic electrochemical
Sweat (2.75 nM–0.04 mM) [54] No
impedance spectroscopy
LOD (–)
(EIS)
0.2–0.6 ng/mL
Saliva Field effect transistor [55] No
LOD 0.005 ± 0.002 ng/mL
2 to 50 ng/mL Electrochemical impedance
Buffer [56] No
LOD 0.66 ng/mL immunoassay
Molecules 2023, 28, 2353 11 of 21

Table 1. Cont.

Sample Source Linear Range, LOD Signal Generation Reference Applicability In Vivo
0.1 to 20 µg/mL Metal-enhanced
Buffer [39] No
LOD 0.02 µg/mL fluoroimmunoassay
Magnetic particle-assisted
Reconstituted 5.0 10−3 and 150 ng mL−1 competitive immunoassay
[57] No
Human serum LOD 3.5 pg mL−1 with differential pulse
voltammetry
Preprocessed 0.1 ß µM to 10 mM Quantum dots fluorescence
[40] No
Saliva LOD 100 pM quenching immunoassay

Apparently, discussed analytical systems are performing with enough accuracy to

measure cortisol concentrations fluctuations with antibodies as biorecognition element
in vitro. Application of such assays to the novel surface substrates and development of
adapted for in vivo signal acquisition strategies may result in CCM device prototype.
Some techniques discussed were developed using antibodies and aptamers (Table 2).

Table 2. Aptasensors.

Sample Applicability
Linear Range, LOD Signal Generation Reference
Source In Vivo
1−256 ng/mL
Electrochemical ZnO
Sweat (2.75 µM–138 µM) [6] No
polymer matrix electrode
LOD (–)
Metalloporphyrin based
50 to 200 nM
Saliva macrocyclic catalyst [58] No
LOD 10 pM
electrochemical sensor
Liquid-ion gated
1 pM to 10 µM
Sweat field-effect transistor [48] No
LOD 10 pM
(FET)

3.3. Molecularly Imprinted Polymer Based Cortisol Biorecognition

A biorecognition element formulated in the presence of the analyte—molecularly
imprinted polymer (MIP)—is another example of nanochemistry used for the cortisol
monitoring task. A self-imprinted polymer functionalized with the appendage steroid—
acrylated cortisol—was created on the single-wall carbon nanotube (SWCNT) surface
interpolymer pockets of the same size and surface charge as a free steroid. As the free
steroid concentration increased, the acrylated appendage cortisol was easily displaced from
the SWCNT surface, releasing the polymer loops farther from the SWCNT. Such flexible
polymer loop movement changes the physico-chemical properties of the sensor to be de-
tected as the analytical signal proportional to the free cortisol concentration (Figure 8) [59].
However, these interpolymer pockets on the surface of single wall carbon nanotubes
(Figure 9) appear to have less specificity to cortisol when performed with a mixture of
steroids—cortisol and cortisone are measured by this system in an indistinguishable man-
ner: the other natural steroids also show some reactivity toward self-imprinting with
cortisol as the appendage polymer. Micromolar sensitivity is not sufficient for a normal
nanomolar range of cortisol concentration in the serum, but the approach shows a dy-
namic response, allowing us to measure the cortisol concentration fluctuations over time,
and could be further modified to produce a more specific response with aan optimized
biorecognition element toward in vivo measurements.
Molecules 2023, 28, 2353 12 of 21

Figure 8. Polymer-bound cortisol hapten (acrylated cortisol) serves as a self-templating element

that creates steroid-specific pockets on the surface of the SWCNT bound polymer loops formed by
the introduction of hydrophobic copolymers anchoring to the SWCNT surface as well as flexible
hydrophilic loops (n, m, k depicts the number of copolymer monomers in each functional segment of
the polymer) ([59] Reprinted with permission from Wiley).

Figure 9. Simplified scheme for the illustration of the equilibrium reaction—displacement of the
interpolymer-bound cortisol hapten (acrylated cortisol) by free cortisol molecules. Competing for
docking closely at the SWCNT surface interpolymer pockets, free cortisol and hapten reconfiguring
polymer loops change the nano-environment of the SWNT, producing the analytical signal based on
the SWNT corona phase recognition measurement to monitor the free cortisol fluctuations over time
in hydrogel, allowing for the free diffusion of steroids from the liquid samples in contact with them
(Reproduced with permission from Wiley [59]).

Lee et al. demonstrated in vivo monitoring with such a SWCNT-doped hydrogel,

but not for cortisol—several steroids were utilized to form a reactive looped imprinted
polymer in a hydrogel, one of which, progesterone, was measured in vivo in mice by
implanting a dialysis bag with an 8 kDa molecular cutoff to demonstrate the applicability
of the CoPhMoRe sensor for the further development of continuous monitoring with such
analytical principles.
Molecules 2023, 28, 2353 13 of 21

Another example of the imprinted polymer for the cortisol monitoring task is the
lossy-mode resonance measured from the partially decladded nanocomposite covered
optical fiber [36]. The 600 µm core silica optical fiber was modified to create a sensing
surface of 15 mm in length. The clad was removed in the waist-like shape of the sensing
region in between the connectors to the light source from one side and the detector from
another intact clad side of the probe. The composite was a 12 nm layer of lossy-mode
resonance generating ZnO, covered with a molecularly imprinted polymer consisting of
pockets of cortisol shape and charge compatible surface interfaces that are formed by
template imprinting and removal prior measurements. The reported sensitivity in the
range of 10−12 –10−6 g/mL of cortisol by the reported technique meets the demands of the
physiological cortisol concentration levels that need to be monitored in (artificial) saliva.
However, the approach requires flow-through of the sample over the decladded sensing
region in the optical fiber connected to the loop-like scheme of optical devices (light source,
detector), making such a measurement setup not quite practical for continuous cortisol
monitoring in vivo, but applicable as a low-volume test system. Molecularly imprinted
polymers are often used as cortisol biorecognition element for cortisol sensing. Analytical
performance of some of such assays is summarized in Table 3.

Table 3. MIP-based sensors.

Sample Applicability In
Linear Range, LOD Signal Generation Reference
Source Vivo
10−12 to 10−6 g/mL Fiber optic Lossy
Artificial
(2.75 pM to 2.75 µM) mode resonance [36] No
saliva
LOD 25.9 fg/mL ZnO/polypyrrole
0.03 to 3.6 × 10−6 g/mL
Electrochemical
Sweat (9.9 µM–82 nM) [60] No
transistor
LOD (–)
Electrochemical
10 ng/mL–60 ng/mL (PDMS doped with
Sweat 27.5 µM–165 µMLOD carbon [61] No
2.0 ng/mL ± 0.4 ng/mL nanotubes-cellulose
crystals)
10 × 10−6 to Yes
Corona phase
100 × 10−6 M (concept
ISF molecular recognition [59]
(10–100 µM) demonstrated on
(CoPhMoRe)
LOD (–) progesterone)
0.5 nM to 64 nM
Saliva Impedimetric sensor [62] No
LOD 0.14 nM

3.4. Alternative Protein Scaffolds for CCM

Alternative protein scaffolds exist and are applicable for cortisol sensing. Naturally,
the glucocorticoid receptor (GR) ligand binding domain (LBD) is a biorecognition element.
The affinity of the LBD of the mineralocorticoid receptor is 0.5 nM, which is typical for
a high affinity cortisol receptor. The affinity of glucocorticoid receptor (GR) to cortisol
was about 5 nM. The binding of cortisol to the LBD of GR results in GR conformational
change, activating its biological activities in the downstream pathway. One of such ac-
tivities of the ligand-bound GR is the binding of the transcriptional co-activators and
co-repressors via specific amino acid motifs. One such co-activator, NCOR1, contains the
LxLL motif, which is specifically bound by allosterically remodeled activated GR LBD
reversely. This biological property of GR was utilized to develop GR LBD and NCOR1
chimeric proteins for intermolecular binding assays based on bio-luminescent probe func-
tion reconstitution [63–65]. As GR LBD binds free cortisol from the sample, it changes
conformationally to obtain affinity to the LxLL motif [66], bending the chimeric protein
in a way that two split luciferase domains are in close proximity, and regains its function
Molecules 2023, 28, 2353 14 of 21

by the substrate reaction emitting fluorescent light, with an intensity proportional to the
cortisol concentration.
Structural rearrangements in GR LBD change its biological properties, allowing the uti-
lization of the switching affinity of GR to other components of transcription activation and
the repression of this nuclear receptor was demonstrated to be useful in the development of
a single-molecule fluorescent probe. This technique was developed by producing a chimeric
protein of two split luciferase domains with a flexible linker in between. This complex is
flanked with GR LBD from one side of the construction and its specific binding protein,
NCOR1 (transcription activator), is the LxLL motif from the other side of the construction
(Figure 10). As GR LBD binds free cortisol from the sample, it changes conformationally to
increase the affinity to the LxLL motif, bending the chimeric protein in the way two split
luciferase domains are in close proximity, and regains its function by the substrate reaction
emitting fluorescent light, with an intensity proportional to the cortisol concentration.

Figure 10. Intermolecular equilibrium binding of the split-luciferase cortisol fluorescent assay for
in vitro cortisol monitoring. As cortisol appears and activates glucocorticoid LBD, which conforma-
tionally changes toward enhanced affinity to the LxLL motif, allowing two luciferase domains to
reconstitute its function and to report a cortisol concentration increase (Reproduced from [65] with
permission from @ACS).

Such proteins were expressed in vitro in the cells and immobilized in wells to func-
tion as a fluorescent test system to perform measurements of cortisol concentration in the
saliva and serum, which demonstrated a high sensitivity and dynamic range, allowing
us to measure the cortisol concentration in the volunteers’ saliva samples with its phys-
iological concentrations of a 10−8 M range with corresponding results when compared
with the ELISA measurement control method. Kim et al. [63–65] demonstrated the sensor
performance across a wide range of concentrations, 10−9 –10−6 M, effectively realizing a
unimolecular biosensor with the intermolecular equilibrium binding signal generation
principle. The authors stated that their cSimgr4 construction was selective only for cortisol
and did not show any response on other steroids.

4. Discussion
The total cortisol levels in plasma varied from 80 to 700 nmol/L in normal subjects
with 90% of cortisol in the plasma protein-bound. In saliva, all cortisol molecules are
free and the concentration ranges from 1.5 to 15 nM [67]. To measure the differences in
1 nM increments of free cortisol, the affinity of the cortisol receptor in the sensor has to
Molecules 2023, 28, 2353 15 of 21

be at least in the nM range. This is comparable to the affinities of monoclonal antibodies

to cortisol. The problem is that ligand binding to the receptor with Kd in nM can be
considered as irreversible in the time intervals of 1 h, which is relevant to physiological
measurements. Hence, the optimal balance between affinity and the half time of dissociation
must be optimized for continuous measurements without chemical “recharging” of the
receptor, as carried out in the salivary sensors by the repetitive acid washing of bound
cortisol. To achieve continuous measurement, several individual recombinant antibodies to
cortisol should be characterized, and the antibodies with optimal Kd based on reasonable
sensitivity and dissociation time selected.Development of a clinically relevant (in terms
of accuracy and rational sample source choice) continuous cortisol sensor has been stated
as the ultimate goal by many research laboratories. However, the technical achievements
of the numerous recent publications have not resolved all of the complex engineering
tasks of an in vivo cortisol monitoring device. It includes an optimal time lag of the
measurements due to the biorecognition reaction kinetics (up to 20 min [3]), allowing us
to distinguish individual cortisol pulses, a proper protection of the sensing surface (from
fouling and matrix background), the identification of the most relevant sample source, and
the biocompatibility of the sensor components for long-term implantation of the sensing
probe for in vivo measurements in the tissue of interest.
The effective membrane protection of such a prospective sensor could again be “bor-
rowed” from the field of glucose sensing, where PSF hollow fiber membranes, designed for
dialysis, are used to contain and protect electrode sensor. One should take into account
the cortisol molecule’s physico-chemical properties, namely, the lipophilic steroid ability
to aggregate and vessel wall stacking in polar solvents. Therefore, CCM device sensor
membrane protection must be developed and optimized to balance the permeability of a
fast rate of cortisol free diffusion but be selective enough to prevent the cortisol-converting
enzymes and matrix from introducing errors into the measurement.
One more example of the cortisol monitoring capable assay publication points to one
of the limitations of the redox and electrochemical biosensors, which is highly dependent
on solution components such as potassium and sodium ions in biofluids, pH, other steroids,
and charged molecules [38]. In order to protect the redox aptasensor, an oil layer has been
introduced to fill the pores of a low-selective track membrane, and thin layers of different
lipids have been studied to assess the cortisol free diffusion rate and sensor fouling over
time [38]. The concept of protecting hydrophobic (lipophilic) analytes such as cortisol is
supported by the results of the published paper of a 5 min response time on concentration
fluctuation with a performance of over 7 h under oil membrane protection from pH 3
outside the sensor compartment.
It is also important to measure the free cortisol concentration inside the circulation in
blood or within ISF, not in excreted fluids such as sweat and saliva. Indeed, only cortisol
concentration in blood provides needed information on complex biological conditions [68,69].
While a substantial progress toward CCM device development has been documented
in numerous publications, a time-resolved, biocompatible, biostable, compact, and effective
biosensor to monitor cortisol rhythm accurately in vivo has not been described yet. A
thorough description of specific stages available in recent publications allow us to predict
some of the crucial features of future CCM analytical systems.
The probe of the sensor must be placed in interstitial fluid or into vascular space. The
latter is more difficult to control and is hardly possible without a professional medical
installation. Interstitial type sensors have been proven to be practical for continuous real-
time glucose measurements. But, the Eversense capsule is still too large, a 18.3 mm of
length and 3.5 mm in diameter, which is implanted with a special needle injector operated
by trained medical personnel. A cortisol sensor will most likely utilize the design of a
microneedle serving as a catheter to deliver a filament containing a modified optical fiber.
Optical fibers are hair-thin and quite flexible and can be durable if it is of a polymer material.
The rationale of measuring cortisol in the interstitial fluid is well-supported in the study
by Venugopal et al. [53]. The electrochemical impedance (EIS) technique and ELISA have
Molecules 2023, 28, 2353 16 of 21

shown a high degree of correlation between ISF and the saliva concentrations of cortisol
across diurnal variation.
It has been reported that approximately 70% higher cortisol numbers are seen in ISF.
The authors concluded that ISF might be a better choice of a bodily fluid to be tested for
levels of cortisol than saliva, serum, or blood. However, most of the published works
on cortisol monitoring in ISF have been performed in vitro by external devices after the
samples of ISF were collected and concentrated.
The interstitial fluid, unlike sweat or saliva, provides the dynamic milieu with constant
flow. The concentration of cortisol in the interstitial fluid follows the fluctuation in the
hormone concentration in the plasma. Cortisol is present in ISF mostly as a free hormone
since the cortisol binding globulin (CBG) is cutoff by the endothelial layer of the vessels.
The receptor (antibody, aptamer or imprinted polymer) has to be optimized in terms
of the thermodynamics of binding (affinity or dissociation constants) and the kinetics of
association/dissociation. This will allow for both the detection of the hormone at a low
concentration and with a sufficiently fast time resolution. An elegant theoretical study was
recently published addressing this problem [69]. The study argued for a pre-equilibrium
biosensor, in which the actual measurement reflects the receptor’s kinetic response and the
algorithm quantifies the changing ligand concentration, analyzing the biosensor output
in the frequency domain, rather than in the time domain. The authors concluded that
antibodies or aptamers with a medium range (micromolar, rather than nanomolar affinity)
should be used in such sensors. A different approach to the solution of the thermodynamics
versus kinetics problem was taken by Lubken et al. [70].
It will be essential for a reversible sensor with high precision to use a single molecular
interaction, as described in [43]. It was shown that a sensor based on an immobilized
monoclonal antibody responded to cortisol in the high nanomolar to low micromolar range
and could monitor cortisol concentrations over multiple hours. The measurement was
made under a stationary microscope through the registration of the Brownian motion of the
particles decorated with the antibody and tethered to the substrate by an extended DNA
bridge. For a wearable sensor, this method must be modified, where the proximity of the
reporter particle to the substrate will generate the optical signal.
Optical fiber-based sensing is a robustly developing area of research. Indeed, micro and
nanofabrication, nanobiotechnological elements, and rationally exploited physico-chemical
and optical properties of sensing waveguides are the variables for the construction of
miniature, biocompatible, and sensitive analytical devices of great promise due to the
utilization of fiber surfaces as assay substrates and signal transducing elements, which can
be easily connected with existing light-guiding commutation hardware.
An optical fiber tip surface, modified with the capture antibody to adsorb S. aureus on
its surface while a secondary antibody in complex with a quantum dot labelled reporter
antibody could be detected by measuring the fluorescence on the fiber surface with a
spectrometer attached to the probe through a bundle-waveguide. A 532 nm excitation laser
light is delivered to the probe tip by the waveguide, and the backscattered QD emission
from the bound S. aureus surface on the optical fiber tip is then read by the spectrometer
attached to the bundle-waveguide [42].
Another example of an optical fiber surface as a sensing probe and assay substrate is
a multimode step-index silica core of 400 µm in diameter, silica in 25 µm TECS cladding,
and a decladded 1-cm tip coated with 45 nm gold to obtain the surface-plasmon resonance
(SPR) substrate. Such an optrode was further modified with aptamers catching HER2, a
cancer biomarker that is the target for the bonding of the monoclonal antibody to HER2.
The optrode is connected by the bundle-waveguide (Y-bundle, FC-PC, and BFT1 connectors
of Thorlabs) to the white light source and the spectrometer reading the backscattered light
(Figure 11) [71].
Molecules 2023, 28, 2353 17 of 21

Figure 11. SPR optrode (a). First fast stage of HER2 capture by Au-bound aptamers. (b) Second long
stage of the amplification of the signal with HER2 specific antibody adsorption on the fiber surface to
enhance the signal 100-fold [71] with permission from Elsevier.

Resonance shift and refractive index changes over time were processed after calibra-
tion, which was performed with two modes of measurement: without amplification (only
HER2 adsorption is detected, with fast reaction kinetics of around 10 min) and ~8 nM
(1 µg/mL) sensitivity, and after ~40 min of incubation, the amplification of the signal
reached 100-fold, as antibodies bound HER3 bound to the fiber surface by aptamers, en-
hancing the signal as the volume of the adsorbed complexes increased. Such amplification
gains showed ~86 pm (10 ng/mL) of sensitivity [71]. Such examples of optic fiber probe
development demonstrate the high practical value of side-emitting and collecting optical
sensors for the development of prospective in vivo monitoring devices.
Taking into account the prospective CCM device analytical performance requirements
and the key features of the considered analytical systems, here, a strategy is proposed to
create a functional CCM prototype.
The metal-enhanced assay previously described in [39] could be applied to optical
fiber sensing and optimized further by using the same approach of intermolecular binding
simply to bend the chimeric protein probe to the surface of the nanostructured optical fiber
probe sensing region. The minimal functional part of GR LBD could be fused through a
flexible protein linker with the LxLL motif.
The addition of a surface-specific tag will be used to immobilize the GR LBD to the
AuNPs on the fiber surface. The sequential covalent attachment of a specific protein tag to
the terminal of the chimeric protein oriented to the outward surface will result in bending
the protein toward the nanostructured metal surface if the GR LBD is activated by free
cortisol (Figure 12).
Binding of the protein probe will shorten the distance between the fluorescent re-
porter and Au nanoparticles on the surface of the fiber probe, which will increase the
fluorescence intensity proportionally to the free cortisol concentration. Introducing such
a multifunctional biorecognition element with a single cortisol binding site and fluores-
cent reporter tethered to the surface must be beneficial for the selectivity of the promi-
nent sensor properties—the binding/dissociation kinetics, noise to signal ratio—and will
probably decrease the technical demands for the protective membrane of the probe for
in vivo measurements.
Molecules 2023, 28, 2353 18 of 21

Such systems, based on the biosensing properties of the natural cortisol receptors
together with a hair-thin, durable, flexible, and reliable surface substrate and signal trans-
ducing element, are promising strategies toward ideal dynamic biosensors for CCM in vivo.

Figure 12. Intermolecular equilibrium binding of the proposed single-molecule metal-enhanced

fluorescent assay on the polymer optical fiber toward the in vivo monitoring of cortisol.

5. Conclusions
A real-time continuous sensor for cortisol remains a desirable but still unreached goal,
even regarding a laboratory prototype. However, there is an understanding of the main
conditions that will allow for such a sensor to be built. All essential elements were already
described and assessed experimentally. This includes a reversible receptor based on the
monoclonal or engineered antibody and immobilized by a flexible bridge, allowing for
a reversible single molecule interaction of the receptor with the analyte. The interaction
of the fluorescent tag with the metal impregnated surface of the optical fiber provides a
signal enhancement. We believe that “corticometers” applicable for laboratory research and
clinical practice will be demonstrated soon, based on the principles outlined in our review.

Author Contributions: V.P.D.—Supervision. P.A.K.—Writing original draft. Y.V.K. and V.P.D.—

Writing and editing. Conceptualization—P.A.K. and Y.V.K. Visualization—P.A.K. All authors have
read and agreed to the published version of the manuscript.
Funding: This research received no external funding.
Institutional Review Board Statement: Not applicable.
Data Availability Statement: Not applicable.
Conflicts of Interest: The authors declare no conflict of interest.
Molecules 2023, 28, 2353 19 of 21

References
1. Bhake, R.C.; Kluckner, V.; Stassen, H.; Russell, G.M.; Leendertz, J.; Stevens, K.; Linthorst, A.C.E.; Lightman, S.L. Continuous Free
Cortisol Profiles—Circadian Rhythms in Healthy Men. J. Clin. Endocrinol. Metab. 2019, 104, 5935–5947. [CrossRef]
2. Chen, L.-S.; Singh, R.J. Niche Point-of-Care Endocrine Testing – Reviews of Intraoperative Parathyroid Hormone and Cortisol
Monitoring. Crit. Rev. Clin. Lab. Sci. 2018, 55, 115–128. [CrossRef]
3. Lee, M.A.; Bakh, N.; Bisker, G.; Brown, E.N.; Strano, M.S. A Pharmacokinetic Model of a Tissue Implantable Cortisol Sensor. Adv.
Healthc. Mater. 2016, 5, 3004–3015. [CrossRef]
4. Boolani, A.; Channaveerappa, D.; Dupree, E.J.; Jayathirtha, M.; Aslebagh, R.; Grobe, S.; Wilkinson, T.; Darie, C.C. Trends in
Analysis of Cortisol and Its Derivatives. In Advancements of Mass Spectrometry in Biomedical Research; Woods, A.G., Darie, C.C.,
Eds.; Advances in Experimental Medicine and Biology; Springer International Publishing: Cham, Switzerland, 2019; Volume 1140,
pp. 649–664, ISBN 978-3-030-15949-8.
5. Madhu, S.; Anthuuvan, A.J.; Ramasamy, S.; Manickam, P.; Bhansali, S.; Nagamony, P.; Chinnuswamy, V. ZnO Nanorod Integrated
Flexible Carbon Fibers for Sweat Cortisol Detection. ACS Appl. Electron. Mater. 2020, 2, 499–509. [CrossRef]
6. Ganguly, A.; Lin, K.C.; Muthukumar, S.; Prasad, S. Autonomous, Real-Time Monitoring Electrochemical Aptasensor for Circadian
Tracking of Cortisol Hormone in Sub-Microliter Volumes of Passively Eluted Human Sweat. ACS Sens. 2021, 6, 63–72. [CrossRef]
[PubMed]
7. Kinnamon, D.; Ghanta, R.; Lin, K.-C.; Muthukumar, S.; Prasad, S. Portable Biosensor for Monitoring Cortisol in Low-Volume
Perspired Human Sweat. Sci. Rep. 2017, 7, 13312. [CrossRef]
8. Persu, A. 11β-Hydroxysteroid Deshydrogenase: A Multi-Faceted Enzyme. J. Hypertens. 2005, 23, 29–31. [CrossRef] [PubMed]
9. Luo, A.; El Gierari, E.T.M.; Nally, L.M.; Sturmer, L.R.; Dodd, D.; Shi, R.-Z. Clinical Utility of an Ultrasensitive Urinary Free
Cortisol Assay by Tandem Mass Spectrometry. Steroids 2019, 146, 65–69. [CrossRef] [PubMed]
10. Harno, E.; White, A. Will Treating Diabetes with 11β-HSD1 Inhibitors Affect the HPA Axis? Trends Endocrinol. Metab. 2010, 21,
619–627. [CrossRef]
11. Markey, K.; Mitchell, J.; Botfield, H.; Ottridge, R.S.; Matthews, T.; Krishnan, A.; Woolley, R.; Westgate, C.; Yiangou, A.;
Alimajstorovic, Z.; et al. 11β-Hydroxysteroid Dehydrogenase Type 1 Inhibition in Idiopathic Intracranial Hypertension: A
Double-Blind Randomized Controlled Trial. Brain Commun. 2020, 2, fcz050. [CrossRef]
12. Molenaar, N.; Groeneveld, A.B.J.; de Jong, M.F.C. Three Calculations of Free Cortisol versus Measured Values in the Critically Ill.
Clin. Biochem. 2015, 48, 1053–1058. [CrossRef] [PubMed]
13. Siiteri, P.K.; Murai, J.T.; Raymoure, W.J.; Kuhn, R.W.; Hammond, G.L.; Nisker, J.A. The Serum Transport of Steroid Hor-
mones. In Proceedings of the 1981 Laurentian Hormone Conference; Elsevier: Amsterdam, The Netherlands, 1982; pp. 457–510;
ISBN 978-0-12-571138-8.
14. Cameron, A.; Henley, D.; Carrell, R.; Zhou, A.; Clarke, A.; Lightman, S. Temperature-Responsive Release of Cortisol from Its
Binding Globulin: A Protein Thermocouple. J. Clin. Endocrinol. Metab. 2010, 95, 4689–4695. [CrossRef] [PubMed]
15. Dhillo, W.; Kong, W.; Le Roux, C.; Alaghband-Zadeh, J.; Jones, J.; Carter, G.; Mendoza, N.; Meeran, K.; O’Shea, D. Cortisol-Binding
Globulin Is Important in the Interpretation of Dynamic Tests of the Hypothalamic–Pituitary–Adrenal Axis. Eur. J. Endocrinol.
2002, 231–235. [CrossRef] [PubMed]
16. Meyer, E.; Nenke, M.; Rankin, W.; Lewis, J.; Torpy, D. Corticosteroid-Binding Globulin: A Review of Basic and Clinical Advances.
Horm. Metab. Res. 2016, 48, 359–371. [CrossRef]
17. Lee, J.H.; Meyer, E.J.; Nenke, M.A.; Falhammar, H.; Torpy, D.J. Corticosteroid-Binding Globulin (CBG): Spatiotemporal Distribu-
tion of Cortisol in Sepsis. Trends Endocrinol. Metab. 2023, S1043276023000139. [CrossRef]
18. Fusani, L. Field Techniques in Hormones and Behavior. In Encyclopedia of Animal Behavior; Elsevier: Amsterdam, The Netherlands,
2019; pp. 488–494; ISBN 978-0-12-813252-4.
19. El-Farhan, N.; Rees, D.A.; Evans, C. Measuring Cortisol in Serum, Urine and Saliva – Are Our Assays Good Enough? Ann. Clin.
Biochem. Int. J. Lab. Med. 2017, 54, 308–322. [CrossRef]
20. McEwen, B.S. Central Effects of Stress Hormones in Health and Disease: Understanding the Protective and Damaging Effects of
Stress and Stress Mediators. Eur. J. Pharmacol. 2008, 583, 174–185. [CrossRef]
21. Jafri, R.Z.; Balliro, C.A.; El-Khatib, F.; Maheno, M.M.; Hillard, M.A.; O’Donovan, A.; Selagamsetty, R.; Zheng, H.; Damiano,
E.R.; Russell, S.J. A Three-Way Accuracy Comparison of the Dexcom G5, Abbott Freestyle Libre Pro, and Senseonics Eversense
Continuous Glucose Monitoring Devices in a Home-Use Study of Subjects with Type 1 Diabetes. Diabetes Technol. Ther. 2020, 22,
846–852. [CrossRef]
22. Colvin, A.E.; Jiang, H. Increased in Vivo Stability and Functional Lifetime of an Implantable Glucose Sensor through Platinum
Catalysis. J. Biomed. Mater. Res. A 2013, 101A, 1274–1282. [CrossRef]
23. Mortellaro, M.; DeHennis, A. Performance Characterization of an Abiotic and Fluorescent-Based Continuous Glucose Monitoring
System in Patients with Type 1 Diabetes. Biosens. Bioelectron. 2014, 61, 227–231. [CrossRef]
24. Irace, C.; Cutruzzolà, A.; Tweden, K.; Kaufman, F.R. Device Profile of the Eversense Continuous Glucose Monitoring System for
Glycemic Control in Type-1 Diabetes: Overview of Its Safety and Efficacy. Expert Rev. Med. Devices 2021, 18, 909–914. [CrossRef]
[PubMed]
25. Parlak, O. Portable and Wearable Real-Time Stress Monitoring: A Critical Review. Sens. Actuators Rep. 2021, 3, 100036. [CrossRef]
26. Riedemann, T.; Patchev, A.V.; Cho, K.; Almeida, O.F. Corticosteroids: Way Upstream. Mol. Brain 2010, 3, 2. [CrossRef] [PubMed]
Molecules 2023, 28, 2353 20 of 21

27. Chung, S.; Son, G.H.; Kim, K. Circadian Rhythm of Adrenal Glucocorticoid: Its Regulation and Clinical Implications. Biochim.
Biophys. Acta BBA Mol. Basis Dis. 2011, 1812, 581–591. [CrossRef] [PubMed]
28. Dahl, R. Systemic Side Effects of Inhaled Corticosteroids in Patients with Asthma. Respir. Med. 2006, 100, 1307–1317. [CrossRef]
[PubMed]
29. Kourakis, S.; Timpani, C.A.; Campelj, D.G.; Hafner, P.; Gueven, N.; Fischer, D.; Rybalka, E. Standard of Care versus New-Wave
Corticosteroids in the Treatment of Duchenne Muscular Dystrophy: Can We Do Better? Orphanet J. Rare Dis. 2021, 16, 117.
[CrossRef]
30. Kapugi, M.; Cunningham, K. Corticosteroids. Orthop. Nurs. 2019, 38, 336–339. [CrossRef]
31. Buttgereit, F.; Doering, G.; Schaeffler, A.; Witte, S.; Sierakowski, S.; Gromnica-Ihle, E.; Jeka, S.; Krueger, K.; Szechinski, J.; Alten, R.
Targeting Pathophysiological Rhythms: Prednisone Chronotherapy Shows Sustained Efficacy in Rheumatoid Arthritis. Ann.
Rheum. Dis. 2010, 69, 1275–1280. [CrossRef]
32. Rensen, N.; Gemke, R.J.; van Dalen, E.C.; Rotteveel, J.; Kaspers, G.J. Hypothalamic-Pituitary-Adrenal (HPA) Axis Suppression
after Treatment with Glucocorticoid Therapy for Childhood Acute Lymphoblastic Leukaemia. Cochrane Database Syst. Rev. 2017,
11, 1465–1858. [CrossRef]
33. Chen Cardenas, S.M.; Santhanam, P.; Morris-Wiseman, L.; Salvatori, R.; Hamrahian, A.H. Perioperative Evaluation and Manage-
ment of Patients on Glucocorticoids. J. Endocr. Soc. 2022, 7, bvac185. [CrossRef]
34. Rison, S.; Rajeev, R.; Bhat, V.S.; Mathews, A.T.; Varghese, A.; Hegde, G. Non-Enzymatic Electrochemical Determination of Salivary
Cortisol Using ZnO-Graphene Nanocomposites. RSC Adv. 2021, 11, 37877–37885. [CrossRef]
35. Goyal, R.N.; Chatterjee, S.; Rana, A.R.S. A Comparison of Edge- and Basal-Plane Pyrolytic Graphite Electrodes towards the
Sensitive Determination of Hydrocortisone. Talanta 2010, 83, 149–155. [CrossRef]
36. Usha, S.P.; Shrivastav, A.M.; Gupta, B.D. A Contemporary Approach for Design and Characterization of Fiber-Optic-Cortisol
Sensor Tailoring LMR and ZnO/PPY Molecularly Imprinted Film. Biosens. Bioelectron. 2017, 87, 178–186. [CrossRef]
37. Rocchitta, G.; Spanu, A.; Babudieri, S.; Latte, G.; Madeddu, G.; Galleri, G.; Nuvoli, S.; Bagella, P.; Demartis, M.; Fiore, V.; et al.
Enzyme Biosensors for Biomedical Applications: Strategies for Safeguarding Analytical Performances in Biological Fluids. Sensors
2016, 16, 780. [CrossRef]
38. Yuan, Y.; DeBrosse, M.; Brothers, M.; Kim, S.; Sereda, A.; Ivanov, N.V.; Hussain, S.; Heikenfeld, J. Oil-Membrane Protection of
Electrochemical Sensors for Fouling- and PH-Insensitive Detection of Lipophilic Analytes. ACS Appl. Mater. Interfaces 2021, 13,
53553–53563. [CrossRef] [PubMed]
39. Safarian, S.M.; Kusov, P.A.; Kosolobov, S.S.; Borzenkova, O.V.; Khakimov, A.V.; Kotelevtsev, Y.V.; Drachev, V.P. Surface-Specific
Washing-Free Immunosensor for Time-Resolved Cortisol Monitoring. Talanta 2021, 225, 122070. [CrossRef]
40. Liu, Y.; Wu, B.; Tanyi, E.K.; Yeasmin, S.; Cheng, L.-J. Label-Free Sensitive Detection of Steroid Hormone Cortisol Based on
Target-Induced Fluorescence Quenching of Quantum Dots. Langmuir 2020, 36, 7781–7788. [CrossRef]
41. Hassan, H.; Bang, O.; Janting, J. Polymer Optical Fiber Tip Mass Production Etch Mechanism to Achieve CPC Shape for Improved
Biosensor Performance. Sensors 2019, 19, 285. [CrossRef] [PubMed]
42. Cui, J.; Zhou, M.; Li, Y.; Liang, Z.; Li, Y.; Yu, L.; Liu, Y.; Liang, Y.; Chen, L.; Yang, C. A New Optical Fiber Probe-Based Quantum
Dots Immunofluorescence Biosensors in the Detection of Staphylococcus Aureus. Front. Cell. Infect. Microbiol. 2021, 11, 665241.
[CrossRef] [PubMed]
43. van Smeden, L.; Saris, A.; Sergelen, K.; de Jong, A.M.; Yan, J.; Prins, M.W.J. Reversible Immunosensor for the Continuous
Monitoring of Cortisol in Blood Plasma Sampled with Microdialysis. ACS Sens. 2022, 7, 3041–3048. [CrossRef] [PubMed]
44. Visser, E.W.A.; Miladinovic, J.; Milstein, J.N. An Ultra-Stable and Dense Single-Molecule Click Platform for Sensing Protein-DNA
Interactions. bioRxiv 2020, 2020.11.25.397737.
45. An, J.E.; Kim, K.H.; Park, S.J.; Seo, S.E.; Kim, J.; Ha, S.; Bae, J.; Kwon, O.S. Wearable Cortisol Aptasensor for Simple and Rapid
Real-Time Monitoring. ACS Sens. 2022, 7, 99–108. [CrossRef] [PubMed]
46. Sekar, M.; Pandiaraj, M.; Bhansali, S.; Ponpandian, N.; Viswanathan, C. Carbon Fiber Based Electrochemical Sensor for Sweat
Cortisol Measurement. Sci. Rep. 2019, 9, 403. [CrossRef] [PubMed]
47. Pennazza, G.; Santonico, M.; Vollero, L.; Zompanti, A.; Sabatini, A.; Kumar, N.; Pini, I.; Quiros Solano, W.F.; Sarro, L.; D’Amico, A.
Advances in the Electronics for Cyclic Voltammetry: The Case of Gas Detection by Using Microfabricated Electrodes. Front. Chem.
2018, 6, 327. [CrossRef] [PubMed]
48. Sheibani, S.; Capua, L.; Kamaei, S.; Akbari, S.S.A.; Zhang, J.; Guerin, H.; Ionescu, A.M. Extended Gate Field-Effect-Transistor for
Sensing Cortisol Stress Hormone. Commun. Mater. 2021, 2, 10. [CrossRef]
49. Zhang, R.; Jia, Y. A Disposable Printed Liquid Gate Graphene Field Effect Transistor for a Salivary Cortisol Test. ACS Sens. 2021,
6, 3024–3031. [CrossRef]
50. Villa, J.E.L.; Garcia, I.; Jimenez de Aberasturi, D.; Pavlov, V.; Sotomayor, M.D.P.T.; Liz-Marzán, L.M. SERS-Based Immunoassay
for Monitoring Cortisol-Related Disorders. Biosens. Bioelectron. 2020, 165, 112418. [CrossRef]
51. Upasham, S.; Tanak, A.; Jagannath, B.; Prasad, S. Development of Ultra-Low Volume, Multi-Bio Fluid, Cortisol Sensing Platform.
Sci. Rep. 2018, 8, 16745. [CrossRef]
52. Kim, Y.-H.; Lee, K.; Jung, H.; Kang, H.K.; Jo, J.; Park, I.-K.; Lee, H.H. Direct Immune-Detection of Cortisol by Chemiresistor
Graphene Oxide Sensor. Biosens. Bioelectron. 2017, 98, 473–477. [CrossRef]
Molecules 2023, 28, 2353 21 of 21

53. Venugopal, M.; Arya, S.K.; Chornokur, G.; Bhansali, S. A Realtime and Continuous Assessment of Cortisol in ISF Using
Electrochemical Impedance Spectroscopy. Sens. Actuators Phys. 2011, 172, 154–160. [CrossRef]
54. Rice, P.; Upasham, S.; Jagannath, B.; Manuel, R.; Pali, M.; Prasad, S. CortiWatch: Watch-Based Cortisol Tracker. Future Sci. OA
2019, 5, FSO416. [CrossRef]
55. Ben Halima, H.; Bellagambi, F.G.; Brunon, F.; Alcacer, A.; Pfeiffer, N.; Heuberger, A.; Hangouët, M.; Zine, N.; Bausells, J.; Errachid,
A. Immuno Field-Effect Transistor (ImmunoFET) for Detection of Salivary Cortisol Using Potentiometric and Impedance
Spectroscopy for Monitoring Heart Failure. Talanta 2022, 123802. [CrossRef]
56. Ben Halima, H.; Zine, N.; Bausells, J.; Jaffrezic-Renault, N.; Errachid, A. A Novel Cortisol Immunosensor Based on a Hafnium
Oxide/Silicon Structure for Heart Failure Diagnosis. Micromachines 2022, 13, 2235. [CrossRef] [PubMed]
57. Moreno-Guzmán, M.; Eguílaz, M.; Campuzano, S.; González-Cortés, A.; Yáñez-Sedeño, P.; Pingarrón, J.M. Disposable Im-
munosensor for Cortisol Using Functionalized Magnetic Particles. The Analyst 2010, 135, 1926. [CrossRef]
58. Fernandez, R.E.; Umasankar, Y.; Manickam, P.; Nickel, J.C.; Iwasaki, L.R.; Kawamoto, B.K.; Todoki, K.C.; Scott, J.M.; Bhansali,
S. Disposable Aptamer-Sensor Aided by Magnetic Nanoparticle Enrichment for Detection of Salivary Cortisol Variations in
Obstructive Sleep Apnea Patients. Sci. Rep. 2017, 7, 17992. [CrossRef]
59. Lee, M.A.; Wang, S.; Jin, X.; Bakh, N.A.; Nguyen, F.T.; Dong, J.; Silmore, K.S.; Gong, X.; Pham, C.; Jones, K.K.; et al. Implantable
Nanosensors for Human Steroid Hormone Sensing In Vivo Using a Self-Templating Corona Phase Molecular Recognition. Adv.
Healthc. Mater. 2020, 9, 2000429. [CrossRef]
60. Parlak, O.; Keene, S.T.; Marais, A.; Curto, V.F.; Salleo, A. Molecularly Selective Nanoporous Membrane-Based Wearable Organic
Electrochemical Device for Noninvasive Cortisol Sensing. Sci. Adv. 2018, 4, eaar2904. [CrossRef] [PubMed]
61. Mugo, S.M.; Alberkant, J. Flexible Molecularly Imprinted Electrochemical Sensor for Cortisol Monitoring in Sweat. Anal. Bioanal.
Chem. 2020, 412, 1825–1833. [CrossRef] [PubMed]
62. Ertuğrul Uygun, H.D.; Uygun, Z.O.; Canbay, E.; Girgin Sağın, F.; Sezer, E. Non-Invasive Cortisol Detection in Saliva by Using
Molecularly Cortisol Imprinted Fullerene-Acrylamide Modified Screen Printed Electrodes. Talanta 2020, 206, 120225. [CrossRef]
63. Kim, S.B.; Sato, M.; Tao, H. Genetically Encoded Bioluminescent Indicators for Stress Hormones. Anal. Chem. 2009, 81, 3760–3768.
[CrossRef] [PubMed]
64. Kim, S.-B.; Paulmurugan, R. Bioluminescent Imaging Systems for Assay Developments. Anal. Sci. 2021, 37, 233–247. [CrossRef]
65. Kim, S.B.; Takenaka, Y.; Torimura, M. A Bioluminescent Probe for Salivary Cortisol. Bioconjug. Chem. 2011, 22, 1835–1841.
[CrossRef] [PubMed]
66. Weikum, E.R.; Knuesel, M.T.; Ortlund, E.A.; Yamamoto, K.R. Glucocorticoid Receptor Control of Transcription: Precision and
Plasticity via Allostery. Nat. Rev. Mol. Cell Biol. 2017, 18, 159–174. [CrossRef]
67. Laudat, M.H.; Cerdas, S.; Fournier, C.; Guiban, D.; Guilhaume, B.; Luton, J.P. Salivary Cortisol Measurement: A Practical
Approach to Assess Pituitary-Adrenal Function. J. Clin. Endocrinol. Metab. 1988, 66, 343–348. [CrossRef] [PubMed]
68. Hammond, G.L. Plasma Steroid-Binding Proteins: Primary Gatekeepers of Steroid Hormone Action. J. Endocrinol. 2016, 230,
R13–R25. [CrossRef] [PubMed]
69. Maganzini, N.; Thompson, I.; Wilson, B.; Soh, H.T. Pre-Equilibrium Biosensors as an Approach towards Rapid and Continuous
Molecular Measurements. Nat. Commun. 2022, 13, 7072. [CrossRef] [PubMed]
70. Lubken, R.M.; Bergkamp, M.H.; de Jong, A.M.; Prins, M.W.J. Sensing Methodology for the Rapid Monitoring of Biomolecules at
Low Concentrations over Long Time Spans. ACS Sens. 2021, 6, 4471–4481. [CrossRef] [PubMed]
71. Loyez, M.; Lobry, M.; Hassan, E.M.; DeRosa, M.C.; Caucheteur, C.; Wattiez, R. HER2 Breast Cancer Biomarker Detection Using a
Sandwich Optical Fiber Assay. Talanta 2021, 221, 121452. [CrossRef]

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual
author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to
people or property resulting from any ideas, methods, instructions or products referred to in the content.